MOSS-Image open text-to-image system targets dense text and layouts
A 5B diffusion transformer plus an agentic prompt module reports 96.3% average text coverage; weights are promised for release.
Stories per day · 30 days
2 RSS
3 stories, newest first · one card per story
Stories tagged #layout
One Figure, Every Canvas: Editable Flowchart Relayout via Agentic Pipeline
An agentic parse-style-layout pipeline relayouts flowcharts across aspect ratios with 68.6% content fidelity.
Mira-Scene: Pixel-Aligned Layouts for Generative 3D Scene
Mira-Scene is a new 3D scene reconstruction framework that improves object layout accuracy using dense canonical coordinate maps and multimodal diffusion techniques.